CN112150434A - Tire defect detection method, device, equipment and storage medium - Google Patents
Tire defect detection method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN112150434A CN112150434A CN202011003217.XA CN202011003217A CN112150434A CN 112150434 A CN112150434 A CN 112150434A CN 202011003217 A CN202011003217 A CN 202011003217A CN 112150434 A CN112150434 A CN 112150434A
- Authority
- CN
- China
- Prior art keywords
- image
- tire
- defect
- detected
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 155
- 238000001514 detection method Methods 0.000 title claims abstract description 68
- 238000003860 storage Methods 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 45
- 238000012545 processing Methods 0.000 claims abstract description 35
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 17
- 230000011218 segmentation Effects 0.000 claims description 19
- 230000009467 reduction Effects 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 6
- 230000000877 morphologic effect Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 230000008030 elimination Effects 0.000 claims description 3
- 238000003379 elimination reaction Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 8
- 230000009286 beneficial effect Effects 0.000 description 7
- 230000003287 optical effect Effects 0.000 description 6
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000005520 cutting process Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000012372 quality testing Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses a method, a device, equipment and a storage medium for detecting tire defects. Wherein, the method comprises the following steps: identifying a belted area of the tire X-ray image by adopting a vertical projection method; processing the belted layer area according to a self-adaptive binarization algorithm to obtain an image to be detected; and determining the tire defect characteristics from the image to be detected according to a preset defect detection model so as to finish the tire defect detection. According to the embodiment of the invention, the image of the belt area of the tire is obtained, the defect detection is carried out on the belt area, and the image to be detected is input into the defect detection model to obtain the detection result. The automatic detection of the defects of the tire belt layers is realized, and the defect detection efficiency is improved.
Description
Technical Field
The embodiment of the invention relates to an image processing technology, in particular to a method, a device, equipment and a storage medium for detecting tire defects.
Background
The tire is used as a key part of an automobile, the structure of an internal cord of the tire is complex, various defects can occur during production, and the driving safety of the automobile is directly or indirectly influenced.
In order to ensure that the quality of the tire is qualified, the defect detection before the tire leaves the factory is an essential link in the tire production process. At present, detecting defects by observing X-ray images of tires by quality testing personnel becomes a conventional detection mode of tire production enterprises, but manual detection wastes manpower and time, the efficiency is low, and the accuracy of detection results is influenced by subjective judgment.
Disclosure of Invention
The embodiment of the invention provides a method, a device, equipment and a storage medium for detecting tire defects, which are used for improving the detection efficiency of the tire defects.
In a first aspect, an embodiment of the present invention provides a method for detecting a tire defect, where the method includes:
identifying a belted area of the tire X-ray image by adopting a vertical projection method;
processing the belted layer area according to a self-adaptive binarization algorithm to obtain an image to be detected;
and determining the tire defect characteristics from the image to be detected according to a preset defect detection model so as to finish the tire defect detection.
In a second aspect, an embodiment of the present invention further provides an apparatus for detecting a tire defect, where the apparatus includes:
the area identification module is used for identifying a belted area of the tire X-ray image by adopting a vertical projection method;
the image obtaining module is used for processing the belted layer area according to a self-adaptive binarization algorithm to obtain an image to be detected;
and the defect determining module is used for determining the tire defect characteristics from the image to be detected according to a preset defect detection model so as to finish the tire defect detection.
In a third aspect, an embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the processor implements the method for detecting a tire defect according to any embodiment of the present invention.
In a fourth aspect, embodiments of the present invention also provide a storage medium containing computer-executable instructions for performing a method of detecting tire defects according to any of the embodiments of the present invention when executed by a computer processor.
According to the embodiment of the invention, the belted area of the tire is obtained through vertical projection, self-adaptive binarization is carried out on the belted area, a binarized image of the belted area is changed according to the difference of the foreground and the background in the belted area, the binarized image to be detected is input into a defect detection model, and the defect characteristics of the tire are output. The problem of among the prior art manual identification tire X-ray image cause the identification precision low is solved, practice thrift manpower and time, realize the self-adaptation of tire defect and detect and automated inspection, improve tire defect detection efficiency and flexibility.
Drawings
FIG. 1 is a schematic flow chart of a method for detecting tire defects according to an embodiment of the present invention;
FIG. 2 is a partial X-ray image of a tire in accordance with a first embodiment of the present invention;
FIG. 3 is a schematic vertical projection view of a tire according to a first embodiment of the present invention;
FIG. 4 is a diagram illustrating a defect detection model according to a first embodiment of the present invention;
FIG. 5 is a block diagram of a tire defect detecting apparatus according to a second embodiment of the present invention;
fig. 6 is a schematic structural diagram of a computer device in a third embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a schematic flow chart of a method for detecting a tire defect according to an embodiment of the present invention, which is applicable to a situation of detecting a tire defect, and the method can be executed by a device for detecting a tire defect, as shown in fig. 1, the method specifically includes the following steps:
and 110, identifying a belt area of the X-ray image of the tire by adopting a vertical projection method.
The belt layer is a material layer which tightly grips the tire body along the circumferential direction of the central line of the tire tread under the tire tread base parts of the radial tire and the belt bias tire, can play a role in alleviating impact, and is an important part for guaranteeing the driving safety. An X-ray image of the tire can be generated by rotationally scanning the inside of the tire through a tire X-ray detection device, fig. 2 is a partial X-ray image of the tire, and a belt region of the X-ray image is identified by a vertical projection method. Because the X-ray image is darker and the gray value is smaller, the X-ray image is not beneficial to observation, the gamma conversion can be carried out on the X-ray image before the belt area is identified, the contrast of the tire image is enhanced, and the identification of the belt area is facilitated. The boundary between the tire sidewall and the tire shoulder is determined by a vertical projection, and fig. 3 is a schematic diagram of the vertical projection of the tire. The number of white pixels on both sides of the dashed line frame in fig. 3 shows the boundaries where the vertical drop is on both sides of the shoulder, respectively, so that the belt region can be divided from the tire X-ray image by vertical projection.
And step 120, processing the banding layer area according to a self-adaptive binarization algorithm to obtain an image to be detected.
The X-ray image in the belt layer area or the X-ray image after gamma change is subjected to binarization processing to realize the belt layer segmentation effect, and the belt layer image after binarization is the image to be detected. Because the gray distribution of the tire belt layer part is not uniform, if the same segmentation threshold value is used for direct binarization, the binarization effect can cause errors of image contents in the belt layer area, and the defect detection precision is influenced. Therefore, the image can be binarized by adopting an adaptive binarization method. The division threshold is a boundary pixel value obtained by binarizing a pixel value, and for example, if a pixel point having a pixel value smaller than 122 is binarized into 0, and a pixel point having a pixel value equal to or larger than 122 is binarized into 255, the pixel value 122 is the division threshold.
In this embodiment, optionally, the processing of the bundling layer region according to the adaptive binarization algorithm includes: dividing the belt area into at least two windows; determining the proportion of each gray-scale pixel in the window according to the number of pixel points of each gray-scale pixel in the window; determining a segmentation threshold of the foreground and the background in the window according to a preset threshold determination algorithm; and carrying out binarization processing on the pixel points of the window according to the segmentation threshold.
Specifically, the belt region is divided into at least two windows, for example, the window size may be s × s pixels, s may be set to one eighth of the image width, and s is the side length of the window. A window may be first divided in the belt region, and after the binarization processing of the window is completed, the window is slid to continue the binarization processing of other parts of the belt region until the binarization processing of the whole belt region is completed. Counting the pixels in the window, and determining the number of pixels of each gray-scale pixel, for example, determining the number of pixels with a pixel value of 0, and determining the number of pixels with a pixel value of 1. Determining the proportion of each gray-scale pixel in all the pixel points of the window, for example, if the number of the pixel points with the pixel value of 0 is 10, and the number of all the pixel points in the window is 100, determining that the proportion of the pixel points with the pixel value of 0 in the window is 10%. And determining the segmentation threshold of the foreground and the background in the window by adopting a preset threshold determination algorithm, which can be an OSTU (extreme Start) algorithm, according to the proportion of each gray-level pixel in the window. The foreground means a portion of the cords in the belt region, and the background means a portion other than the cords in the belt region. Determining the number of pixels in the foreground, and determining the proportion of the number of foreground pixels in the pixel points in the window image according to the number of the pixels in the foreground; and determining the number of pixel points in the background, and determining the proportion of the number of the background pixel points in the window image according to the number of the pixel points in the background. The average gray values of the foreground and background in the window, and the total average gray value of the window, are determined. Calculating to obtain the variance of the foreground image and the background image, performing vertical projection on the window image to obtain a vertical projection histogram, and selecting the gray value of the vertical projection histogram corresponding to the maximum variance as the segmentation threshold of the window. The variance is calculated as follows:
u＝w0×u0+w1×u1；
g＝w0×(u0-u)2+w1×(u1-u)2；
the two formulas are combined to obtain:
g＝w0×w1×(u0-u1)2；
wherein, w0The ratio of the number of foreground pixels to the number of pixels in the window image, w1The number of background pixels is the proportion of the number of pixels in the window image, u0Is the average gray level of the foreground in the window, u1Is the average gray level of the background in the window, g is the variance of the foreground image and the background image, and u is the total average gray level of the window image.
After the segmentation threshold is obtained, binarization processing is performed on the pixel points in the window, the pixel points with the pixel values smaller than the segmentation threshold are set as 0, and the pixel points with the pixel values equal to or larger than the segmentation threshold are set as 255. After the binarization processing of the window is finished, the window is slid to the other part of the belt area, and the calculation of the variance and the determination of the segmentation threshold value are continuously carried out until the binarization processing of the whole belt area is finished. The beneficial effects of the arrangement are that the binarization operation is carried out on the windows according to the segmentation threshold value of each window, the self-adaptive binarization processing adapts to the condition that the gray distribution of the belted layer part of the tire is uneven, the belted layer boundary in the belted layer area can be distinguished conveniently, and the binarization effect is obviously improved.
In this embodiment, optionally, processing the bundling layer region to obtain an image to be detected, further includes: performing noise elimination on the belt layer area after binarization processing by adopting morphological operation, and determining a belt layer boundary of the belt layer area; and dividing the belt layer area according to the belt layer boundary in the belt layer area to obtain an image to be detected.
Specifically, after the belt region of the binarization process is obtained, there is a possibility that noise may exist in the processed image, and therefore, the morphological filtering operation may be performed on the binarized image again. The morphology operation can eliminate redundant noise and divide the belted layer, and the morphology operation is adopted to carry out morphology filtering on the binarized image to eliminate the noise, so that a clearer belted layer boundary is obtained. Noise generated by binarization can be eliminated through 3 times of morphological closed operation, and the final belt layer boundary is determined. And dividing the belted layer area according to the belted layer boundary in the belted layer area to obtain an image to be detected in the belted layer area. The beneficial effects of the arrangement are that the boundary of the belted layer is more accurate through morphological operation, the influence of image noise on the detection of the belt defect is avoided, and the precision of the detection of the belt defect of the tire is improved.
And step 130, determining tire defect characteristics from the image to be detected according to a preset defect detection model so as to complete tire defect detection.
The image to be detected is input into a preset defect detection model, and the defect detection model can be used for identifying tire defects and classifying the tire defects to obtain a detection result of the tire defects.
In this embodiment, optionally, the defect detection process of the preset defect detection model includes: acquiring an image to be detected; overlapping the characteristic images output by at least two convolution layers to obtain an overlapped image; and inputting the superposed image into a network output layer to obtain a target frame of the tire defect in the image to be detected and a defect classification result.
Specifically, a Convolutional neural Network based on deep learning may be adopted, for example, a fast Region-based Convolutional Network method (fast Convolutional Network method) model is adopted as a defect detection model of a tire belt layer, and a VGG16 Network is adopted as a Convolutional layer for feature extraction. The VGG16 may include at least two convolutional layers, and the image to be detected is input into a preset defect detection model, so as to obtain feature images output by a plurality of convolutional layers, for example, there are five convolutional layers, so as to obtain feature images of the last three convolutional layers, and the output feature images include the feature image of the last convolutional layer. The plurality of characteristic images are superposed to obtain a superposed image, and the characteristics extracted by different convolution layers can exist on the same image. The multiple characteristic images can be adjusted according to the preset size requirement, and the images with the same size can be overlapped. And inputting the superposed image into a network output layer of the defect detection model, outputting a target frame of the tire defect of the image to be detected, and determining the position of the tire defect and a defect classification result. The defect detection model can be preset with images of different defects and defect names, and the detected defect images can be compared with the preset defect images to determine the names of the detected defects. The beneficial effect who sets up like this lies in, uses the convolutional neural network to carry out the detection and the discernment of defect, reduces artifical process that detects, practices thrift manpower and time, avoids the manual work to the subjective judgement influence result accuracy of defect, improves defect detection's efficiency and precision.
In this embodiment, optionally, the step of superimposing the feature images output by the at least two convolution layers to obtain a superimposed image includes: adjusting second characteristic images output by other convolutional layers according to the size of the first characteristic image output by the last convolutional layer; wherein the other convolutional layers are convolutional layers except the last convolutional layer in the at least two convolutional layers; and overlapping the first characteristic image and the adjusted second characteristic image to generate an overlapped image.
Specifically, the number of feature images output by the convolutional layer is at least two, and one of the feature images is the feature image output by the last convolutional layer. The characteristic image output by the last layer of the convolution layer is a first characteristic image, and the characteristic images output by the convolution layers except the last layer of the convolution layer are second characteristic images. And adjusting the size of the second characteristic image based on the size of the first characteristic image to make the size of the second characteristic image consistent with that of the first characteristic image. For example, feature images of the last three convolutional layers may be fused, feature images of three different dimensions are changed into the same size by ROI stacking (Region of Interest Pooling layer), and the feature images of the same size are fused in an overlapping manner, so as to obtain an overlapped image. The beneficial effects of the arrangement are that after the image to be detected is subjected to multilayer convolution, the characteristic extraction is carried out step by step, the unobvious tire defects in the image are gradually blurred, and the characteristics in each characteristic image are gathered on the same image through the superposition of the multilayer characteristic images, so that the omission of the defects is avoided, and the precision and the efficiency of the defect detection are improved.
When the feature images output by at least two convolution layers are superposed to obtain a superposed image, the candidate frame of the first feature image can be extracted by adopting an area recommendation network.
Specifically, when the superimposed image is obtained, the first feature image output by the last layer of convolutional layer may be input to an RPN (Region recommendation Network). The convolution can be performed by adopting a VGG16 network, and the characteristic image output by the last layer of convolution layer is the output image of VGG 16. And extracting candidate frames of the first characteristic image through an RPN network, wherein the first characteristic image and the superposed image have the same size, the image contents can be mapped with each other, and the candidate frames of the superposed image are obtained according to the candidate frames of the first characteristic image, so that the superposed image can be subjected to defect detection, and the tire defect detection efficiency is improved.
In this embodiment, optionally, the step of inputting the superimposed image into the network output layer to obtain a target frame of the tire defect in the image to be detected and a defect classification result includes: reducing the dimension of the superposed image by using the convolution layer; inputting the superposed image subjected to the dimensionality reduction into a network output layer, and outputting a target frame of a tire defect image in the image to be detected; and determining the defect type of the tire defect image in the target frame according to the preset incidence relation between the tire defect type and the tire defect image.
Specifically, after the superimposed image is obtained, a convolution layer can be used for reducing the dimension of the superimposed image, so that the dimension of the superimposed image conforms to the input dimension of the network output layer, the network output layer is used for outputting the detection result of the tire defect, and the target frame of the position of the tire defect in the image to be detected can be output. The incidence relation between the tire defect type and the tire defect image can be preset, and the defect type matched with the tire defect image in the target frame is searched according to the detected target frame. The beneficial effect who sets up like this lies in, through the convolution dimensionality reduction, avoids the unable circumstances of confirming the testing result of network output layer, and can carry out automatic identification to the defect classification, practices thrift the time that the manpower detected, effectively improves defect detection's efficiency.
FIG. 4 is a schematic diagram of a defect detection model. The feature extraction module 401 has five convolutional layers, the last three convolutional layers are used for outputting respective feature images to the feature image module 402, the first feature image output by the last convolutional layer obtains a candidate frame through the RPN network module 403, the size of the feature image is adjusted in the feature image module 402, a superimposed image is generated in the superimposed module 404, and the candidate frame of the superimposed image can be determined according to the candidate frame generated by the RPN network module 403. The superimposed image 404 is sent to a convolution dimensionality reduction module 405, the convolution dimensionality reduction module 405 may perform dimensionality reduction using a convolution with a scale of 1, so that the dimensionality of the superimposed image is consistent with the dimensionality of an image that may be received by the network output layer 406, and the convolution dimensionality reduction module 405 uses a convolution of 1 × 1. After the dimension reduction is performed on the superimposed image, the superimposed image after the dimension reduction is input to the network output layer module 406, and the network output layer module 406 may output the image at the tire defect position, label the tire defect position with the target frame to display the tire defect position and the image at the tire defect position, and display the classification result of the tire defect. For example, a tire defect may be a crease, cross or break in the cords, or the like.
Before the defect detection model is used, a defect data sample set can be constructed, and the Faster R-CNN network can be trained. The defect sample data set can be horizontally, vertically or specularly inverted, and the number of samples is increased. And adding a defect marking frame to the sample image in the sample set, and marking the defects in the sample image. And cutting the sample images in the defect sample data set at the height of 1200 pixels to enable each sample image to comprise at least one defect marking frame, and cutting one sample image into a plurality of training samples in a sliding cutting mode. Training the Faster R-CNN model by using the processed training samples can be realized under a deep learning framework, the learning rate is set to be 0.001, the maximum iteration number is 15 ten thousand, and finally the defect detection model is obtained.
According to the technical scheme of the embodiment, a belted layer area of the tire is obtained through vertical projection, self-adaptive binarization is carried out on the belted layer area, a binarized image of the belted layer area is changed according to the difference of the foreground and the background in the belted layer area, the binarized image to be detected is input into a defect detection model, and the defect characteristics of the tire are output. The problem of among the prior art manual identification tire X-ray image cause the identification precision low is solved, practice thrift manpower and time, realize self-adaptation detection and automated inspection to tire defect, improve tire defect detection efficiency and flexibility.
Example two
Fig. 5 is a block diagram of a tire defect detection apparatus according to a second embodiment of the present invention, which is capable of executing a tire defect detection method according to any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method. As shown in fig. 5, the apparatus specifically includes:
the region identification module 501 is used for identifying a belt region of a tire X-ray image by adopting a vertical projection method;
an image obtaining module 502, configured to process the band-constrained layer region according to a self-adaptive binarization algorithm to obtain an image to be detected;
and a defect determining module 503, configured to determine a tire defect feature from the to-be-detected image according to a preset defect detection model, so as to complete tire defect detection.
Optionally, the image obtaining module 502 includes:
a window dividing unit for dividing the belt region into at least two windows;
the pixel proportion determining unit is used for determining the proportion of each gray-scale pixel in the window according to the number of pixel points of each gray-scale pixel in the window;
the segmentation threshold determining unit is used for determining segmentation thresholds of the foreground and the background in the window according to a preset threshold determining algorithm;
and the binarization processing unit is used for carrying out binarization processing on the pixel points of the window according to the segmentation threshold value.
Optionally, the image obtaining module 502 further includes:
the noise elimination unit is used for eliminating noise of the belt layer area after binarization processing by adopting morphological operation and determining the belt layer boundary of the belt layer area;
and the to-be-detected image determining unit is used for dividing the belt layer area according to the belt layer boundary in the belt layer area to obtain the to-be-detected image.
Optionally, the sub-module of the preset defect detection model and the defect detection process include:
the image acquisition submodule is used for acquiring the image to be detected;
the image superposition submodule is used for superposing the characteristic images output by the at least two convolution layers to obtain a superposed image;
and the result output submodule is used for inputting the superposed image into the network output layer to obtain a target frame of the tire defects in the image to be detected and a defect classification result.
Optionally, the image superimposition sub-module is specifically configured to:
adjusting second characteristic images output by other convolutional layers according to the size of the first characteristic image output by the last convolutional layer; wherein the other convolutional layers are convolutional layers except the last convolutional layer in the at least two convolutional layers;
and overlapping the first characteristic image and the adjusted second characteristic image to generate an overlapped image.
Optionally, the result output submodule is specifically configured to:
reducing the dimension of the superposed image by using the convolution layer;
inputting the superposed image subjected to the dimensionality reduction into a network output layer, and outputting a target frame of a tire defect image in the image to be detected;
and determining the defect type of the tire defect image in the target frame according to the preset incidence relation between the tire defect type and the tire defect image.
According to the embodiment of the invention, the belted area of the tire is obtained through vertical projection, self-adaptive binarization is carried out on the belted area, a binarized image of the belted area is changed according to the difference of the foreground and the background in the belted area, the binarized image to be detected is input into a defect detection model, and the defect characteristics of the tire are output. The problem of among the prior art manual identification tire X-ray image cause the identification precision low is solved, practice thrift manpower and time, realize the self-adaptation of tire defect and detect and automated inspection, improve tire defect detection efficiency and flexibility.
EXAMPLE III
Fig. 6 is a schematic structural diagram of a computer device according to a third embodiment of the present invention. FIG. 6 illustrates a block diagram of an exemplary computer device 600 suitable for use in implementing embodiments of the invention. The computer device 600 shown in fig. 6 is only an example and should not bring any limitations to the function and scope of use of the embodiments of the present invention.
As shown in fig. 6, computer device 600 is in the form of a general purpose computing device. The components of computer device 600 may include, but are not limited to: one or more processors or processing units 601, a system memory 602, and a bus 603 that couples various system components including the system memory 602 and the processing unit 601.
The system memory 602 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)604 and/or cache memory 605. The computer device 600 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 606 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 6, commonly referred to as a "hard drive"). Although not shown in FIG. 6, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 603 by one or more data media interfaces. Memory 602 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
A program/utility 608 having a set (at least one) of program modules 607 may be stored, for example, in memory 602, such program modules 607 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. The program modules 607 generally perform the functions and/or methods of the described embodiments of the invention.
The computer device 600 may also communicate with one or more external devices 609 (e.g., keyboard, pointing device, display 610, etc.), with one or more devices that enable a user to interact with the computer device 600, and/or with any devices (e.g., network card, modem, etc.) that enable the computer device 600 to communicate with one or more other computing devices. Such communication may occur via an input/output (I/O) interface 611. Moreover, the computer device 600 may also communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet) via the network adapter 612. As shown in FIG. 6, the network adapter 612 communicates with the other modules of the computer device 600 via the bus 603. It should be appreciated that although not shown in FIG. 6, other hardware and/or software modules may be used in conjunction with the computer device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
The processing unit 601 executes various functional applications and data processing by running the program stored in the system memory 602, for example, implementing the method for detecting a tire defect provided by the embodiment of the present invention, including:
identifying a belted area of the tire X-ray image by adopting a vertical projection method;
processing the banding layer area according to a self-adaptive binarization algorithm to obtain an image to be detected;
and determining the tire defect characteristics from the image to be detected according to a preset defect detection model so as to finish the tire defect detection.
Example four
The fourth embodiment of the present invention further provides a storage medium containing computer executable instructions, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for detecting a tire defect provided by the fourth embodiment of the present invention, and the method includes:
identifying a belted area of the tire X-ray image by adopting a vertical projection method;
processing the banding layer area according to a self-adaptive binarization algorithm to obtain an image to be detected;
and determining the tire defect characteristics from the image to be detected according to a preset defect detection model so as to finish the tire defect detection.
Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or computer device. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A method of detecting a tire defect, the method comprising:
identifying a belted area of the tire X-ray image by adopting a vertical projection method;
processing the belted layer area according to a self-adaptive binarization algorithm to obtain an image to be detected;
and determining the tire defect characteristics from the image to be detected according to a preset defect detection model so as to finish the tire defect detection.
2. The method according to claim 1, wherein processing the belt region according to an adaptive binarization algorithm comprises:
dividing the belt area into at least two windows;
determining the proportion of each gray-scale pixel in the window according to the number of pixel points of each gray-scale pixel in the window;
determining segmentation threshold values of the foreground and the background in the window according to a preset threshold value determination algorithm;
and carrying out binarization processing on the pixel points of the window according to the segmentation threshold.
3. The method of claim 2, wherein processing the belt region to obtain an image to be detected further comprises:
performing noise elimination on the belt layer area after binarization processing by adopting morphological operation, and determining a belt layer boundary of the belt layer area;
and dividing the belt layer area according to the belt layer boundary in the belt layer area to obtain an image to be detected.
4. The method according to claim 1, wherein the defect detection process of the preset defect detection model comprises:
acquiring the image to be detected;
overlapping the characteristic images output by at least two convolution layers to obtain an overlapped image;
and inputting the superposed image into a network output layer to obtain a target frame of the tire defects in the image to be detected and a defect classification result.
5. The method of claim 4, wherein superimposing the feature images output by the at least two convolutional layers to obtain a superimposed image comprises:
adjusting second characteristic images output by other convolutional layers according to the size of the first characteristic image output by the last convolutional layer; wherein the other convolutional layers are convolutional layers except the last convolutional layer in the at least two convolutional layers;
and overlapping the first characteristic image and the adjusted second characteristic image to generate an overlapped image.
6. The method according to claim 4, wherein inputting the superimposed image to a network output layer to obtain a target frame of the tire defect and a defect classification result in the image to be detected comprises:
reducing the dimension of the superposed image by using a convolution layer;
inputting the superposed image subjected to the dimensionality reduction into a network output layer, and outputting a target frame of a tire defect image in the image to be detected;
and determining the defect type of the tire defect image in the target frame according to the preset incidence relation between the tire defect type and the tire defect image.
7. A tire defect detection apparatus, comprising:
the area identification module is used for identifying a belted area of the tire X-ray image by adopting a vertical projection method;
the image obtaining module is used for processing the belted layer area according to a self-adaptive binarization algorithm to obtain an image to be detected;
and the defect determining module is used for determining the tire defect characteristics from the image to be detected according to a preset defect detection model so as to finish the tire defect detection.
8. The apparatus of claim 7, wherein the image acquisition module comprises:
a window dividing unit for dividing the belt region into at least two windows;
the pixel proportion determining unit is used for determining the proportion of each gray-scale pixel in the window according to the number of pixel points of each gray-scale pixel in the window;
the segmentation threshold determining unit is used for determining segmentation thresholds of the foreground and the background in the window according to a preset threshold determining algorithm;
and the binarization processing unit is used for carrying out binarization processing on the pixel points of the window according to the segmentation threshold.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the program, implements the method of detecting tire defects according to any one of claims 1 to 6.
10. A storage medium containing computer executable instructions for performing the method of detecting tire defects according to any one of claims 1 to 6 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011003217.XA CN112150434A (en) | 2020-09-22 | 2020-09-22 | Tire defect detection method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011003217.XA CN112150434A (en) | 2020-09-22 | 2020-09-22 | Tire defect detection method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112150434A true CN112150434A (en) | 2020-12-29 |
Family
ID=73896283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011003217.XA Pending CN112150434A (en) | 2020-09-22 | 2020-09-22 | Tire defect detection method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112150434A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113066079A (en) * | 2021-04-19 | 2021-07-02 | 北京滴普科技有限公司 | Method, system and storage medium for automatically detecting wood defects |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107437243A (en) * | 2017-06-19 | 2017-12-05 | 中国科学院自动化研究所 | Tire method for detecting impurities and device based on x-ray image |
| CN110120036A (en) * | 2019-04-17 | 2019-08-13 | 杭州数据点金科技有限公司 | A kind of multiple dimensioned tire X-ray defect detection method |
| CN110660049A (en) * | 2019-09-16 | 2020-01-07 | 青岛科技大学 | Tire defect detection method based on deep learning |
-
2020
- 2020-09-22 CN CN202011003217.XA patent/CN112150434A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107437243A (en) * | 2017-06-19 | 2017-12-05 | 中国科学院自动化研究所 | Tire method for detecting impurities and device based on x-ray image |
| CN110120036A (en) * | 2019-04-17 | 2019-08-13 | 杭州数据点金科技有限公司 | A kind of multiple dimensioned tire X-ray defect detection method |
| CN110660049A (en) * | 2019-09-16 | 2020-01-07 | 青岛科技大学 | Tire defect detection method based on deep learning |
Non-Patent Citations (3)
| Title |
|---|
| 张斌;林森;高书征;: "基于图像处理的轮胎X光图像杂质检测技术", 橡塑技术与装备, no. 09 * |
| 艾笑天: "基于轮胎X光图像的缺陷自动检测算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, vol. 2019, no. 08, pages 1 - 86 * |
| 郑修楠: "基于X光成像和纹理分析的轮胎带束层缺陷检测", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, vol. 2020, no. 01, pages 1 - 64 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113066079A (en) * | 2021-04-19 | 2021-07-02 | 北京滴普科技有限公司 | Method, system and storage medium for automatically detecting wood defects |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110276754B (en) | Surface defect detection method, terminal device and storage medium | |
| CN114842017B (en) | HDMI cable surface quality detection method and system | |
| CN109241867B (en) | Method and device for recognizing digital rock core image by adopting artificial intelligence algorithm | |
| US11686703B2 (en) | Automated analysis of analytical gels and blots | |
| CN111861990B (en) | Method, system and storage medium for detecting bad appearance of product | |
| CN111415339B (en) | Image defect detection method for complex texture industrial product | |
| CN116485779B (en) | Adaptive wafer defect detection method and device, electronic equipment and storage medium | |
| CN113870202A (en) | Far-end chip defect detection system based on deep learning technology | |
| CN112017154A (en) | Ray defect detection method based on Mask R-CNN model | |
| CN113947597B (en) | Industrial defect detection method, device and medium based on shielding reconstruction | |
| CN117392042A (en) | Defect detection method, defect detection apparatus, and storage medium | |
| CN114841974A (en) | Nondestructive testing method and system for internal structure of fruit, electronic equipment and medium | |
| CN112150434A (en) | Tire defect detection method, device, equipment and storage medium | |
| CN111914706B (en) | Method and device for detecting and controlling quality of text detection output result | |
| JP2020135051A (en) | Fault inspection device, fault inspection method, fault inspection program, learning device and learned model | |
| CN116309561A (en) | Surface defect identification method based on anti-creeping insulating material | |
| CN113763384B (en) | Defect detection method and defect detection device in industrial quality inspection | |
| CN113516328B (en) | Data processing method, service providing method, device, equipment and storage medium | |
| CN115393589A (en) | Universal DCS process flow chart identification conversion method, system and medium | |
| CN115345806A (en) | Object detection method and device, electronic equipment and readable storage medium | |
| KR20230036650A (en) | Defect detection method and system based on image patch | |
| Lin et al. | Automatic detection of shoulder bending defects in tire X-ray images | |
| CN113420716B (en) | Illegal behavior identification and early warning method based on improved Yolov3 algorithm | |
| KR101646838B1 (en) | Counting method for track images | |
| CN111724352B (en) | Patch LED flaw labeling method based on kernel density estimation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |